araT5-Base-with-LoRA

This model is a fine-tuned version of UBC-NLP/AraT5v2-base-1024 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.1185
• Bleu: 12.5314
• Rouge: 0.5045
• Gen Len: 14.066

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Rouge	Gen Len
4.2656	1.0	7500	2.6346	9.6457	0.4263	14.0196
3.2885	2.0	15000	2.3718	10.7953	0.4683	14.1228
3.0225	3.0	22500	2.2546	11.4633	0.4853	14.0032
2.865	4.0	30000	2.1520	12.0114	0.4999	13.9988
2.747	5.0	37500	2.1185	12.5314	0.5045	14.066

Framework versions

• PEFT 0.12.0
• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1